Collaborative QEIB Research: Spatio-temporal Modeling of Species Distributions and Biodiversity at High Resolution - Integrating Population and Climate Responses

了解地球上物种的分布和生物多样性的模式是生态学和生物地理学的核心问题。这个问题的关键在于理解现在、过去和未来的气候如何在空间和时间上驱动生物模式。该项目将重点发展贝叶斯分层统计模型，将模式和过程整合到多个层面上，从生物种群到物种再到区域气候。世界正处于气候变化的特殊阶段，这可能会导致物种分布模式发生巨大变化。许多科学家指出，一场迫在眉睫的保护危机。南非提供了一个案例研究，这一点尤其正确。然而，目前手头的工具不足以提供足够的预测能力（及其不确定性）和足够的空间分辨率来理解、规划和解决危机。本研究将集中于预测和解释南非开普植物区（Cape Floristic Region， CFR）的植物物种分布（特别是变形科），利用植物分布和表现的数据为区域层面的气候降尺度提供信息。它将利用气候数据来了解物种如何对气候变化做出反应。CFR是一个物种多样性和地理分化程度极高的地区，有着详细的气候和物种分布记录。将开发新的统计模型来预测和解释物种丰度和种群水平的表现，并改进对最近过去和不久将来气候的预测。对于这个系统的特定过程所获得的洞察力，以及我们开发的模型和工具，将广泛适用于理解世界上其他地方的其他系统。本项目为培养生态学、气候学、统计学、建模等学科的学生和博士后提供了一个独特的机会。它将为服务不足的学生制定积极的招生和留校计划，并将与美国和南非建立联系，开展一项以生物多样性和气候变化问题为重点的K-12教育推广计划。

Understanding the distribution of species and patterns of biodiversity across the earth is a central problem in ecology and biogeography. Key to this problem is understanding how present, past and future climate may drive biological patterns spatially and temporally. This project will focus on developing Bayesian hierarchical, statistical models that integrate pattern and process across many levels, from populations of organisms to species to regional climates. The world is in an extraordinary phase of climate change that will drive potentially dramatic changes in species distribution patterns. Many scientists have pointed to a looming conservation crisis. Southern Africa provides a case study where this is especially true. Yet the tools currently at hand are inadequate for providing sufficient predictive power (and uncertainties thereof) with sufficient spatial resolution to understand, plan for and address the crisis. This study will concentrate on predicting and explaining plant species distributions (especially for the Proteaceae family) in the Cape Floristic Region (CFR) of South Africa, using data on plant distributions and performances to inform the downscaling of climate at a regional level. It will use the climate data to learn how species respond to changing climate. The CFR is a region with exceptionally high diversity and levels of geographic differentiation and with detailed climate and species distribution records. Novel statistical models will be developed to predict and explain species abundances and population-level performances, and refine forecasts for recent past and near future climates. The insight gained on the processes specific to this system, as well as the models and tools we develop, will be broadly applicable to understanding other systems elsewhere in the world. This project provides a unique opportunity for trans-disciplinary training of students and post-doctoral associates in ecology, climatology, statistics, and modeling. It will developed an active recruiting and retention program for underserved students, and will cultivate a K-12 educational outreach program, with US-South African linkages, that focuses on issues of biodiversity and climate change.